RealWire Limited
Delivering Relevance, Releasing Influence
Issue Date

Release Feed
RSS FeedActix Feed
Share Release
Tweet Facebook LinkedIn
blink it Reddit! Stumble It! Add to Diigo Bookmark this on Delicious
Tweet Facebook LinkedIn
Press Release

Actix One platform passes stringent enterprise loading tests

Actix sets industry leading 23,408 Transactions per Second in the OPTS-N network engineering load benchmark at the 1000 concurrent user’s level

London, 15 December 2008 - Actix, the leading supplier of Network Status Management (NSM) software and systems, today announced its Actix One platform (made generally available on 1 December 2008) has passed tough independent third party scalability tests using the OPTS-N benchmark. The tests, run by Software Quality Services (SQS), Hewlett Packard and Intel, prove the platform’s suitability for use in the largest and most critical telecoms grade NSM applications.

The Optimization Transactions-N or OPTS-N benchmark is a rigorous test of NSM solutions, which aims to verify scalability and absolute software performance suitability for deployments by the world’s largest mobile network operators.

“The benchmark is particularly demanding as it combines multiple simultaneous batch data processing activities with real time network analysis while supporting up to 1000 concurrent interactive users and maintaining consistent system response times within preset user guidelines. All user measurements are made to a web browser interface and involve extensive graphics, making this a very serious test of the application and requires exceptional software engineering to achieve” said Richard Kateley, Actix Chief Technology Officer.

Conducted by SQS, a leading independent provider of system and software testing and quality improvement services, Hewlett Packard and Intel, the tests utilized HP’s Load Runner product to provide accurate user loads and response-time measurements and took place at the HP/ Intel Solution Centre in Grenoble, France, over several weeks during September and October 2008.

“Chief Technology Officers at Tier 1 operators now know for certain that they can retire the multitude of disparate, non-integrated engineering tools and replace them with a single enterprise suite solution, dramatically cutting network OPEX and knowing for sure they will experience no adverse scalability effects for their users in network engineering and operations,” said Alex Hawker, CEO of Actix. “The results of these tests are an important benchmark, demonstrating Actix One’s ability to withstand even the harshest processing loads very elegantly. This degree of scalability is a core capability for any company wishing to implement a truly automated network optimization function.”

The Actix One system was configured with an Oracle RDBMS R.10G and architected as a pure web application with three tier SOA architecture covering data acquisition, storage & analysis, visualization and optimization. A total of 96 Intel Xeon CPU cores and 5Tb of data storage were deployed across the entire system, with an HP c7000 series blade chassis.

Key results for the tests include:-
• Exceeding previous processing benchmarks by significant margins for live processing of terabyte network measurement data loads
• Capable of supporting 200,000 network sectors and 1,000 concurrent users.
• Large performance margins and up to 1,000 users at the limit (100% achievement) of specified response times
• Providing diagnostic and self-optimizing responses to network issues in less than 10% of specified customer target times
• Handling both daily and near real-time diagnosis and optimization tasks as much as 90% faster than customer target times

Live commercial deployments of Actix One are already capable of processing more than a terabyte of raw network measurement data per day, and supporting wireless networks with more than 200,000 radio sectors. Actix One is the most widely deployed system in this field, and has been shown to deliver cost reduction benefits of tens of millions of dollars when deployed. The system delivers large-scale automation of radio network engineering processes, creating more efficient and effective processes - crucial at a time when network OPEX is under intense pressure. Actix One is now deployed with Tier 1 operators and vendors on every major continent.

- - -

About Actix
With increasingly fierce competition and sky-high customer expectations, mobile operators around the world are striving to extract every last bit of performance and value from one of their biggest assets, the radio access network.  UK-based Actix helps wireless carriers enhance revenue growth and improve operating margins by delivering intelligent, automated network performance engineering and network status management systems.  Actix solutions feature embedded wireless expertise, automating key processes and enabling very significant efficiencies in the deployment and operation of carriers’ networks, delivering measurably higher performance at lower cost.  Over 10,000 engineers from 227 operators – including Verizon, O2, AT&T and T-Mobile – in 106 countries globally depend upon Actix software every day to help improve coverage for more than 1,100,000,000 subscribers. For more information, please see www.actix.com

About OPTS-N
The Optimization Transactions-N (OPTS-N) benchmark simulates a real world network engineering and operations environment.  The performance metric reported by OPTS-N reflects multiple aspects of the capability of the system to process network optimization transactions, including query processing power and transaction throughput for multiple concurrent users. The benchmark uses real rather than simulated customer network data and specifies a daily engineering cycle covering 16 daily and 34 real-time analytical and self-optimizing processing loads, all to run concurrently with 5 measures for end user response times. OPTS-N exercises a range of system capabilities, including:
- Ability to simultaneous execute a variety of transaction types of differing complexity
- Multiple on-line concurrent, interactive user sessions
- Targeted system and application execution time
- VLSDBs consisting a wide variety of sizes, attributes, and relationships

Output is the number of engineering transactions per second achieved for a specified number of concurrent users (1000) 100% of which must experience response times within the defined limits.

As an example, a scenario running 1000 Radio Engineers, 100,000 Sectors, 50 analitic Rules in 3.56 minutes = 50 x 100,000 / (3.56 * 60) = 23,408 OPTS-N.

Press contacts:
Nicola Smart/Sarah Rhodes
AxiCom (for Actix)
Email: actix@axicom.com
Tel: +44 (0)20 8392 4050